Learner experiences
What practitioners say after completing the programmes
Accounts from people who have worked through the courses — what they found useful, what was difficult, and where they've applied what they learned.
Back to Home240+
Practitioners enrolled since 2023
4.7
Average satisfaction score (out of 5)
62%
Of Eng. course learners go on to MLOps track
18
Cohorts completed across all programmes
What learners say
A selection of reviews from recent cohorts
Pattarapol Phitchayapan
Data Engineer · Bangkok
The Software Engineering course covered things I had been doing informally for years but had never properly understood. The version control module alone changed how I structure experimental branches. The exercises were actually difficult in the right way — not contrived.
April 2025 · Software Engineering for AI
Napatsawan Thianthong
ML Engineer · Chiang Mai
I did the MLOps track after reading through the course structure carefully. The office hours made a bigger difference than I expected — having an instructor who actually runs ML pipelines in production meant the answers were specific rather than textbook. The sixteen weeks felt like the right length.
March 2025 · MLOps and Production Track
Supakorn Kanchanarak
Senior Developer · Bangkok
I was skeptical about whether an online format could deliver the depth I needed for the Architecture Programme. The small-group reviews changed that. You can't hide in a group of five people, which means the feedback is genuine. My portfolio piece ended up being something I use as reference at work.
April 2025 · AI Systems Architecture Programme
Wannipa Bunnak
Research Analyst · Bangkok
Coming from a research background, I had the analysis skills but not the engineering ones. The Software Engineering course helped me understand why my notebooks kept breaking in other people's environments. I appreciated that the pace was consistent and the exercises were cumulative.
February 2025 · Software Engineering for AI
Thanakorn Rattanawong
Software Engineer · Phuket
The MLOps track moved at a pace that worked for someone with a full-time job. I watched the recorded sessions on evenings and worked through the exercises on weekends. The office hours were useful for asking about things that came up in my actual work, not just the course material.
March 2025 · MLOps and Production Track
Araya Promjit
Tech Lead · Bangkok
The Architecture Programme is the first course I've done that asked me to produce a document I'd actually show to a colleague. The structure meant I had to think carefully about trade-offs, not just absorb concepts. The cohort reviews were sometimes uncomfortable, which is exactly what I needed.
January 2025 · AI Systems Architecture Programme
In more depth
Three learner journeys
Case study 01 · Software Engineering for AI
From exploratory scripts to a shared, tested project
The situation
A Bangkok-based data analyst had built several useful analysis tools as standalone scripts. When a colleague tried to run them, nothing worked. The tools had dependencies that weren't documented, paths that were hardcoded, and no tests.
What changed
During the ten-week course, the analyst restructured the primary tool as a proper Python package with documented dependencies, a test suite for the main transformation functions, and a CI configuration that ran those tests on each commit.
The outcome
The restructured project was adopted by the team as a shared tool. Two colleagues have since contributed improvements. The analyst enrolled in the MLOps track the following quarter to work on deploying the tool's output as an API.
"The exercises in week three and four changed how I think about project structure. I've restructured almost everything I've written since." — W.B., Research Analyst
Case study 02 · MLOps and Production Track
Moving a classification model from notebook to a stable internal service
The situation
An ML engineer at a logistics company had a well-performing classification model that ran as a batch process every night. The model had drifted over three months without anyone noticing, because there was no monitoring on its output distribution.
What changed
The MLOps track provided a structured way to think about the operational side: experiment versioning, model packaging, deployment patterns, and — most importantly — basic observability. The engineer implemented a monitoring check on the model's prediction distribution by week eleven of the track.
The outcome
The monitoring caught a distribution shift six weeks after implementation. The model was retrained and redeployed before the issue reached the downstream report it fed. The engineer is now the team's reference point for deployment questions.
"The observability module was the one I didn't think I needed. It turned out to be the most directly useful thing in the track." — N.T., ML Engineer
Case study 03 · AI Systems Architecture Programme
Documenting a multi-model system for stakeholder review
The situation
A tech lead was responsible for a system that combined three separate models with different data flows and evaluation cycles. When asked to produce a technical proposal for extending the system, she had no clear framework for how to represent the system's current state or proposed changes.
What changed
The Architecture Programme's focus on system boundaries and interface design gave her a vocabulary and a set of diagramming conventions for representing the existing system clearly. The portfolio project she chose was a written architecture document for the existing system plus a proposed evaluation pipeline.
The outcome
The document was accepted as the technical reference for the extension project. The lead used it in four stakeholder reviews over the following two months. The evaluation pipeline she proposed was built by a junior engineer using the specification she had written.
"I learned how to describe a system so that someone else could build parts of it without asking me every question. That was the skill I actually needed." — A.P., Tech Lead
Get in touch
Contact details
Phone
+66 2 384 5712Address
56 Soi Aree 4, Phaya Thai
Phaya Thai, Bangkok 10400
Thailand
Office Hours
Monday – Friday: 09:00 – 18:00 ICT
Saturday: 10:00 – 14:00 ICT
Professional standing
Recognitions and affiliations
DEPA Contributing Provider
Recognised by Thailand's Digital Economy Promotion Agency as a contributing provider in the country's AI skills development effort, January 2025.
Thai Data Science Community
Active member of the Bangkok data and ML practitioner community. Instructors regularly contribute to meetups and technical writing within the network.
Registered Thai entity
Garuda Tech operates as a registered company in Thailand, issues VAT-compliant invoices, and maintains formal enrolment records for all cohorts.
Join the next cohort
Find out which programme is right for where you are
Send an enquiry and we'll respond within one working day with information on the next available cohort.
Send an Enquiry